Splat | WiseTome.com

When we try to recursively understand the working of each piece of code, we come to a point where the software gradually moves to the world of hardware.  The transistors form the notable, if not the only important unit of this hardware world - the realm of electronics.

Let us abstain from dwelling into the history of transistors.  It had a journey of its own from vacuum tubes to the solid-state devices or the semiconductors.  And more over, this realm of electronics is the home for number of equations that try and model the behavior. They are a headache we can avoid when we try to understand how the transistors do their part the workings of a software system.

The structure of transistors is composed of three parts - Emitter, Base and Collector.  Again, we will steer clear of the actual physical structure or the mathematical models.  To understand the origins of these names would require us to go into the history and look at how the vacuum tubes performed.

In a nut shell, we will look at how the transistor is used as a switch, which would then be used to perform logic operations (NOT, AND, OR, XOR).

In the most common setup of a transistor in an electronic circuit, signal is applied between Base and Emitter, an output is tapped out of Emitter and Collector.  The behavior of the transistor can be understood by plotting a graph of this output against a given input.

As you may notice, from a zero input signal, to a definite amount of signal, the output is some where close to zero.  This area is the ‘cut-off’ region of the transistor.  Next, as the input is increase, the output signal would increase by a factor.  (This factor, any radio hobbyist, would mention as amplification factor).  The amplification region is not of much interest to our scope of logic gates.  If the input signal is increased after a particular point, the output remains a constant high signal.  This is called the ’saturation’ region of transistor.

When a transistor is used only in the cut-off and the saturation regions, it is said that the ‘transistor is used as a switch’.  The input signal controls the output current to be high or low - it acts as an electronically controlled switch.

This switch mode of transistor is used in logic gates.  Logic gates may be formed using different setups using combinations of transistors with other electronic units like resistors, and diodes.  Depending on the units used, gates can be said to belong to a family.

• Diode Logic - No transistors are used. A simpler device called diode can be used to create different logic gates, but they have their limitations.
• Resistor Transistor Logic - As the name suggests, resistors and transistors are used for the logic gates.
• Diode Transistor Logic - Diodes and transistors are used.
• Transistor Transistor Logic - Using transistors alone to perform the logic operations in the gates became an industry standard when the circuits were integrated to for chips.
• Emitter Coupled Logic - These are transistor setups that are configured to operate at high speeds.
• CMOS Logic - The normal transistors called BJT - Bipolar Junction Transistor, had the notoriety of drawing large amount of power.  Complementary Metal Oxide Semiconductors (CMOS) overcame the hurdle, by performing the switching action while drawing less power.

This is the list of some of the milestones and prominent families of transistors and other electronic devices that make up logic gates. Note that there are numerous variations, and it would be hard to stuff it all into one article.

We will look at the basics of logic gates in detail, in the next article of this series.